Rysz

https://forums.unraid.net/topic/60217-plugin-nut-v2-network-ups-tools/page/80/#findComment-1587572 https://forums.unraid.net/topic/60217-plugin-nut-v2-network-ups-tools/page/82/#findComment-1589967

Brigading and rallying up people with one-sided accusations creates a hostile environment on the forums. I have hidden the previous messages pending staff review, they can still be seen by moderators and staff members. Please understand that this topic is not the appropriate channel to appeal any bans from the Unraid Discord. All such bans can be appealed by creating a support ticket through the Unraid support system instead - thank you! Any further posts of this nature will be hidden, pending review by a staff member.

I'm curious, do you know why this happens (in technical terms) ?

Nice, the thanks is all to @Wu23333 !

The update with the patch is now available, please let me know if it works for you now @Wu23333 @minkshaman .

That is great news, thanks a lot for your effort, I'll package a patched version and we'll use that for the plugin! I hope it will also resolve the issue for @minkshaman with the next update of the plugin.

@Wu23333 @minkshaman Can you try running with -vvvvvv, like `sg_ses -vvvvvv --all --json --no-time /dev/sg12? Maybe it prints some information leading up to the segmentation fault that gives me some clues...

Sorry, same applies to you, we're out of luck as the sg_ses binary is both OS-embedded and the sg3_utils project is unmaintained. If anyone has experience with GDB and could provide me with a stack trace, I might be able to patch it up, but without a crashing device I can't do anything.

Unfortunately the format is wildly different, to the point it'd require a rewrite of the core daemon logic. It's not something I can do at the moment, but I'll put it on the TO-DO list for the future... (no promises, I'm afraid). The issue is that your device doesn't fully implement the SES standard and sends malformed responses to some commands. This in turn is not handled gracefully by the - unfortunately unmaintained - sg_ses program, which crashes with the unexpected response. Even if I were to identify where the crash in the sg_ses program is coming from and fixed it, I couldn't ship it as it's embedded in the Unraid OS. My plugin and daemon are basically reliant on the sg_ses program working normally, which it should if the device has a clean SES implementation. I appreciate you posting the files though, and I hope to make use of them to design an alternative polling path for such devices in the future - many thanks! 🤗

Thanks for the feedback, I'd say this is definitely a coincidence and I'd shutdown the system and RMA the memory, good you caught it early-ish before it caused more problems.

I agree with this and it's less reliant on information from the UPS during the outage.

I've added a custom wipe method for pre-clearing Unraid disks as part of the latest update. This is not to replace the official plugin, but to offer an alternative for power users used to and liking nwipe. I'm hoping to (maybe) eventually get this included in the original nwipe program, but for that it needs more testing. While I've tested it personally, please report successes or failures adding nwipe pre-cleared disks to the array here. A core motivation was wishing to pre-clear on non-Unraid systems, so there's also a multi-architecture Docker container: https://github.com/desertwitch/nwipe-pc/pkgs/container/nwipe-pc The container allows to prepare new disks without putting/wiping them in a production system, or to keep pre-cleared spares. You could now theoretically even do pre-clears on a Raspberry Pi with the disk connected over USB... if you're patient enough! 🫠

I'm pretty sure this is the battery that's no longer able to hold the charge, 30% is quite low. Many older batteries voltage will collapse under load even when showing sufficient charge. You can test this by e.g. connecting a lamp or some other non-critical appliance and see when the UPS really loses power vs. the charge it says. If could also be a low charge threshold that kills the power, if there is such a setting (if the UPS has a screen, check the manual). But I highly suspect this is the battery and Unraid cannot fully shutdown before the voltage collapses.

That's great, can you run both pages with --json and post the output in one file for each page? Maybe I can put something together with that or find the bug in the sg_ses tool.

No, it should work if your disks work normally in Unraid - it's possible that the in-band SES protocol is not fully compliant for the enclosure.

I'm afraid nothing we can do then, it's most likely a malformed/unknown response from the enclosure. The segmentation fault is due to an unhandled exception in the sg_ses binary (which is unfortunately unmaintained). Thanks for trying though and reporting back, even if it did not work out in the end...

What happens when you run: sg_ses --all --json --no-time /dev/sg25

I've updated the version to the latest release: https://github.com/trapexit/mergerfs/releases/tag/2.41.0 Please ideally unmount all pools before updating and read carefully the changelog linked above.

The service just wraps the sg_ses binary, which should support all SES-capable devices (including yours). I think the issue is that your device does not support the timestamp it tries to get from it and sends a malformed response. I've added an argument which avoids this, as we do not strictly need the timestamp, and the latest update should fix it - let me know!

Wie mehrfach gesagt, aktuell bestehen technische Probleme - bitte einfach später nochmal probieren. Nachdem wir uns hier aber immer mehr im Kreis zu drehen scheinen, schließe ich das Thema mal soweit. Bei anhaltenden Problemen bitte gerne jederzeit Kontakt mit dem Team aufnehmen: https://unraid.net/support Generell können wir bei Zahlungs- oder Rechnungsproblemen als Community eher wenig weiter helfen... sorry! 🙂

Unabhängig davon gibt es aktuell Probleme bei der Anzeige mancher Zahlungsanbieter. Ist aber aktuell in Klärung, sollte also hoffentlich bald mehr als nur die Kreditkarte anzeigen.

Die Zahlungsmethoden sind aktuell in Überarbeitung, Lastschrift (Giropay/EPS) ist aber definitiv geplant meines Wissens. Das System wurde erst vor ein paar Tagen umgestellt, würde also in ein paar Tagen nochmal reinschauen, dann sind die hoffentlich bereits aktiviert.

And this was with the UPS connected via USB? Nothing is detected at all... which leads me to think USB port, USB cable...? I'd start with trying a USB 2.0 port, if you have one, or any other USB port really... Especially as Eaton are among the best supported and most used devices with NUT.

Please post the NUT debug package that you can find in NUT settings. Just to clarify, you want to connect to a remote NUT server, or the UPS is connected to your Unraid server?

Thanks, you can now switch the NUT backend to "preview (latest build)" before updating. After the update you'll be on the version that has this change baked in, backend should say something including "master". If you saw this after updating already, switch the NUT backend to "preview (latest build)" and reinstall the plugin or just reboot.